- observium - lists.observium.org

Incomplete device name (bug?)
by Richard Hartensveld 28 Aug '13

28 Aug '13

Hi all, If I add a device with a hostname of br0.abcdefg , it will show up nicely on the front page, however, when I add a device with a hostname of br0.abcdefgh (or longer), the device's name will be 'br0' on the front page under "device alerts" and "Recent Eventlog Entries". If you hoover over the device name, he popup screen wil display the correct device name. Could this be a bug, or am I doing something wrong ? Thanks. gr, Richard

2 1

Re: [Observium] OS Recognition
by Adam Armstrong 28 Aug '13

28 Aug '13

I am in the process of migrating this to the os script... :) Sander Steffann <sander(a)steffann.nl> wrote: >> Hi I just wanted to ask if observium is suppose to detect the flavour of linux or is this the correct behaviour? > >I think you want this part of the manual: >http://www.observium.org/wiki/Linux_Distribution_Detection > >Cheers :-) >Sander > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Re: [Observium] Mis-aligned data in memcached app
by Adam Armstrong 28 Aug '13

28 Aug '13

Seems like a good idea. Adam. Sander Steffann <sander(a)steffann.nl> wrote: >Hi, > >When debugging why my memcached graphs weren't working I discovered the following problem: > >SQL[SELECT * FROM `applications` WHERE `device_id` = '41' AND `app_type` = 'memcached'] >Including: applications/memcached.inc.phpmemcachedmemcached(127.0.0.1:11211) >SQL[SELECT app_id FROM `applications` WHERE `device_id` = '41' AND `app_instance` = '127.0.0.1:11211'] >RRD[cmd[update /opt/observium/rrd/v25.invalpool.nl/app-memcached-127.0.0.1_11211.rrd N:64:2.0.16-stable:0:0:5:1045:6202:8:0:0:14045:129043:6202:0:0:0:0:7184655:13093279] >stdout[ERROR: /opt/observium/rrd/v25.invalpool.nl/app-memcached-127.0.0.1_11211.rrd: conversion of '2.0.16-stable' to float not complete: tail '.16-stable'] >stderr[]] > >Obviously the data in the rrdtool update command were wrong. A version number shouldn't be in there... So I looked at what the memcached agent script was returning and how that was parsed by the poller memcached module. It turns out that there is a difference. The agent script returns a list of values in this order: > >accepting_conns, auth_cmds, auth_errors, bytes, bytes_read, bytes_written, cas_badval, cas_hits, cas_misses, cmd_flush, cmd_get, cmd_set, cmd_touch, conn_yields, connection_structures, curr_connections, curr_items, decr_hits, decr_misses, delete_hits, delete_misses, evicted_unfetched, evictions, expired_unfetched, get_hits, get_misses, hash_bytes, hash_is_expanding, hash_power_level, incr_hits, incr_misses, libevent, limit_maxbytes, listen_disabled_num, pid, pointer_size, reclaimed, reserved_fds, rusage_system, rusage_user, threads, time, total_connections, total_items, touch_hits, touch_misses, uptime, version > >The poller module parses them in this order: > >accepting_conns, auth_cmds, auth_errors, bytes, bytes_read, bytes_written, cas_badval, cas_hits, cas_misses, cmd_flush, cmd_get, cmd_set, conn_yields, connection_structures, curr_connections, curr_items, decr_hits, decr_misses, delete_hits, delete_misses, evictions, get_hits, get_misses, incr_hits, incr_misses, limit_maxbytes, listen_disabled_num, pid, pointer_size, rusage_system, rusage_user, threads, time, total_connections, total_items, uptime, version > >As you can see this breaks at cmd_touch: the agent sends a variable the poller doesn't expect, and as the data is sent value-only it has no way to determine that the imput is in different places than it expects... > >What I suggest to do is the following: >- update the agent so that key:value is sent instead of just value >- update the poller module so that it can parse key:value input >- make the poller module also accept the current value-only input as a fall-back > >If we agree that this is the way forward I'll send in a patch. > >Cheers, >Sander > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

OS Recognition
by David Peall 28 Aug '13

28 Aug '13

Hi I just wanted to ask if observium is suppose to detect the flavour of linux or is this the correct behaviour? Linux ns1.coza.net.za 3.2.0-30-generic #48-Ubuntu SMP Fri Aug 24 16:52:48 UTC 2012 x86_64 Hardware Generic x86 64-bit Operating System Linux 3.2.0-30-generic Regards -- David Peall

2 2

Mis-aligned data in memcached app
by Sander Steffann 28 Aug '13

28 Aug '13

Hi, When debugging why my memcached graphs weren't working I discovered the following problem: SQL[SELECT * FROM `applications` WHERE `device_id` = '41' AND `app_type` = 'memcached'] Including: applications/memcached.inc.phpmemcachedmemcached(127.0.0.1:11211) SQL[SELECT app_id FROM `applications` WHERE `device_id` = '41' AND `app_instance` = '127.0.0.1:11211'] RRD[cmd[update /opt/observium/rrd/v25.invalpool.nl/app-memcached-127.0.0.1_11211.rrd N:64:2.0.16-stable:0:0:5:1045:6202:8:0:0:14045:129043:6202:0:0:0:0:7184655:13093279] stdout[ERROR: /opt/observium/rrd/v25.invalpool.nl/app-memcached-127.0.0.1_11211.rrd: conversion of '2.0.16-stable' to float not complete: tail '.16-stable'] stderr[]] Obviously the data in the rrdtool update command were wrong. A version number shouldn't be in there... So I looked at what the memcached agent script was returning and how that was parsed by the poller memcached module. It turns out that there is a difference. The agent script returns a list of values in this order: accepting_conns, auth_cmds, auth_errors, bytes, bytes_read, bytes_written, cas_badval, cas_hits, cas_misses, cmd_flush, cmd_get, cmd_set, cmd_touch, conn_yields, connection_structures, curr_connections, curr_items, decr_hits, decr_misses, delete_hits, delete_misses, evicted_unfetched, evictions, expired_unfetched, get_hits, get_misses, hash_bytes, hash_is_expanding, hash_power_level, incr_hits, incr_misses, libevent, limit_maxbytes, listen_disabled_num, pid, pointer_size, reclaimed, reserved_fds, rusage_system, rusage_user, threads, time, total_connections, total_items, touch_hits, touch_misses, uptime, version The poller module parses them in this order: accepting_conns, auth_cmds, auth_errors, bytes, bytes_read, bytes_written, cas_badval, cas_hits, cas_misses, cmd_flush, cmd_get, cmd_set, conn_yields, connection_structures, curr_connections, curr_items, decr_hits, decr_misses, delete_hits, delete_misses, evictions, get_hits, get_misses, incr_hits, incr_misses, limit_maxbytes, listen_disabled_num, pid, pointer_size, rusage_system, rusage_user, threads, time, total_connections, total_items, uptime, version As you can see this breaks at cmd_touch: the agent sends a variable the poller doesn't expect, and as the data is sent value-only it has no way to determine that the imput is in different places than it expects... What I suggest to do is the following: - update the agent so that key:value is sent instead of just value - update the poller module so that it can parse key:value input - make the poller module also accept the current value-only input as a fall-back If we agree that this is the way forward I'll send in a patch. Cheers, Sander

1 0

No graphic image after the svn update
by waqar 27 Aug '13

27 Aug '13

hi all i just updated my observium using svn. after the update the graph image is not displaying at all.. i quickly revert back to my orignal version that is 4280 but still not graph image . i have deleted one of my device and then add it back but still no graph image.. ps: to be clear i m talking about no graph image.

1 1

Location debug : how ?
by Xavier Beaudouin 27 Aug '13

27 Aug '13

Hello, For some reason, on my observium at work, seems that location doesn't want to work. On device menun, still marked as "unkown". I tried to find is there is an issue with the location part, but on a setup at home, it is working. Location like Paris, FR / Belfort, FR still fail into unknown, so I wish to debug what happened, and how to fix it (maybe this an issue with my setup, but I cannot find where). Any hints about that? Regards, Xavier -- Xavier Beaudouin - Senior Network and System Administrator Infrastructure and network director Horizon Software - http://www.hsoftware.com/ 13 rue La Fayette - 75009 PARIS - France Phone: +33 (0)1 4260 9490 Fax: +33 (0)1 44 56 97 01

3 5

Horizontal Scalability
by Ciro Iriarte 24 Aug '13

24 Aug '13

Hi!, anybody considered scaling adding more servers instead of going to a bigger one?. Using something like Open Grid Scheduler for example. Currently I'm maxing out our server (2 x Xeon E5-2630, 16 cores) with 37500 ports, 548 devices cplanning:~ # uptime 13:19pm up 151 days 3:03, 2 users, load average: 43.37, 43.00, 42.60 Regards, CI.-

4 7

Apps incremental graph
by andres fariña 22 Aug '13

22 Aug '13

Hi All, Currently we are using app features to collect stats from our mobile platforms as troughput, for example Transactions Per Second. Is there an easy way to convert incremental RRD values to Transaction Per Second? Thanks for the help. Regards, Andres

1 0

how to support observium
by Chris 22 Aug '13

22 Aug '13

Hi there, I'm very new to observium and I set up a lot of devices now to be monitored. As I'm managing a big network running on EIGRP I'm missing some things I formerly built in cacti "per hand". I had always an eye on EIGRP queries, EIGRP Topo routes, EIGRP updates and such things. Is it possible to support you to bring in some new graphs into observium? unfortunately I'm not an php-coder but I've got long-term experience in networking and snmp. Nice features where also: - Usage of ISDN-Channels on voice-routers/Gateways (cisco) - Working GFX on MGE-UPS (we've got a big Powerware-Galaxy and want to see output power, battery runtime, and so on) - Working GFX on SmartUPS (older RM1400 - APS OS 3.7.3) (I#m missing output load) - More Sensors monitored of cisco 4500s. and I think I will have many more Ideas ;-) Greetings and many thanks for this great piece of software! Chris

1 0